Contribution of prosodic and paralinguistic

cues to the translation of evidentiary audio

The International Journal for
recordings
Translation & Interpreting
Research
trans-int.org
Raymond Chakhachiro
Western Sydney University
r.chakhachiro@westernsydney.edu.au

DOI: 10.12807/ti.108202.2016.a04  

Abstract: This study examines accuracy in the translation and transcription of

evidentiary audio recordings in the Australian context. Verbatim translation
requested by crime agencies and courts is investigated and translation and
transcription methods are suggested with reference to conversation analysis.
The purpose of evidentiary audio recordings dictates a faithful translation;
however, the prevalent ‘written to be read’ translation and transcription styles
used by crime agencies can jeopardise the output, given the problems created
in reflecting the speakers’ intentions, moods, power and attitudes. The
credibility of transcripts when tendered in evidence in court hinges on the
quality of the translation. In addition to the stylistic accuracy of the translation
of speakers’ interactions, the present paper argues that important discursive
information exhibited in the suprasegmental features in conversation should
be documented on transcripts, including prosodic and paralinguistic elements,
such as intonation, timing of responses and volume. When strategically used,
these features can help in placing the last pieces of the jigsaw puzzle, and
producing ‘audible’, ‘written to be read as if spoken’ texts.

Keywords: Evidentiary audio recordings, transcription and translation,

conversation analysis

1. Introduction

Translation of audio recordings is the transfer of meaning from one spoken

language to another, which involves the transcription or conversion of speech
into a written text. It is a unique field and mode of translation, which is
concerned with speech-types having their specific purposes that dictate the
translation and transcription approach to be adopted. Audio materials
requiring translation and/or transcription include monolingual recordings for
research purposes (e.g., Gumperz, 1996; Sacks, 1995), recorded statements
(Edwards, 1995; Teichman, 2000) and bilingual material for contrastive
analyses (e.g., Bolinger, 1989).
Evidentiary audio recording (EAR) is the spoken material recorded by
crime agencies, using various methods, such as listening devices and
telephone interception, to track suspected criminal activity. This material may
be used as evidence in criminal or civil litigation. Where EAR material is
spoken fully or partially in a foreign language, e.g., Arabic in Australia, its
transcription involves translation into the official or default language of the

Translation  &  Interpreting  Vol  8  No  2  (2016) 46

country where the material is to be used – English in this context. The
transcription and translation (translation for short) of conversation spoken in
another language is an established, specialised sub-field of court ‘interpreting’
(Edwards, 1995). However, despite the potential implications of translations
tendered in evidence, and the likelihood of their being scrutinised by the
defence and rigorously contested in cross-examination, the scarce literature on
the translation of EAR hovers around technicalities and presentation (e.g.,
Edwards, 1995) rather than the pertinent notion of accuracy. This study
attempts to address this deficiency and contribute to a further understanding of
the meaning of accuracy in the translation of evidentiary recordings.

1.1 Deficiencies of the prevalent translation practice

In the prevalent translation practice, conversation is treated as written text.
Grammar, syntax, structure and vocabulary in conversations in languages
other than English (LOTEs) are usually standardised, which obscures or
distorts the interlocutors’ intentions and sociological information embedded in
style and register.
Having the foreign language transcribed alongside its English translation
(cf. Edwards, 1995), without having systematic recourse to and documentation
of the conversation’s prosodic and paralinguistic cues, is marginally useful
when it comes to refreshing the memory of the translator or the bilingual
reviser of the original in cross-examination. However, this exercise runs the
risk of treating the original as a written text in the translation process and in
court.
The predominant verbatim translation or, at best, ‘semantic translation’
(Newmark, 1988), devoid of proper encoding of prosody, may provide
meaningless or ambiguous output when it lends itself to more than one
interpretation and thus may give rise to endless arguments in court among
litigants.
A functional-pragmatic translation approach (House, 2001) or
communicative translation method (Newmark, 1988) – subject to clarity of the
recording, and availability of context and co-text – take into consideration the
prosodic features in the listening process but without having them actually
documented. This could attract legitimate argument by the defence and often
be deemed inadmissible in court on grounds of subjective interpretation. What
is more, the principle of accuracy of court interpreting based on pragmatic
considerations (Berk-Seligson, 2002; Hale, 1996) does not assist the translator
of EAR, who, in addition to the potential lack of access to context and
background knowledge, has no recourse to feedback, repetition, clarification,
and, most importantly, the kinetic cues of conversation.
Unlike interpreting, translation and transcription proper, and despite the
demand for and importance of EAR translation, the field is acutely under-
researched and lacks technical and ethical standards. A freelance translator can
be forced to follow totally different sets of in-house translation guidelines
depending on the agency. In the absence of commonly agreed-upon
guidelines, translators for the same agency can have different understandings
of the translation method required due to the lack of systematic training. This
situation becomes more serious when the defence engages an independent
reviser as expert witness.
EAR materials can be riddled with exophoric references (Shlesinger,
1994), idiosyncratic ellipsis (i.e., grammatical cohesive devices used as part of
the interlocutors’ shared knowledge (cf. Halliday & Hasan, 1976)), coded and
telepathic language, and unintelligible items. These inherent peculiarities of
EAR lead to calling into question the adequacy and consistency of the
prevalent translation method, which thus motivates the present study.

Translation  &  Interpreting  Vol  8  No  2  (2016) 47  

2. Speech type and verbatim translation

The source language of EAR is talk-in-interaction, and thus bears all the
stylistic hallmarks of natural speech, including the use of dialectal or
colloquial words or expressions, abbreviated forms, ellipsis and prosodic
features, as well as complementary non-verbal visual means such as
accompanying physical gestures. In natural spoken language, parts of
utterance and discourse meanings are communicated through style, non-verbal
cues, and voice. According to Crystal (1997, p. 171), pitch and loudness are
“the source of the main linguistic effects”, which, along with effects “arising
out of the distinctive use of speed and rhythm, are collectively known as the
prosodic features of language”.
In this paper, the source language is dialectal and generally characterised
as informal/illiterate at the crossroad between colloquial Arabic (cf.
Versteegh, 2001) and Modern Standard Arabic (MSA), the official written and
formal spoken language in the Arab world. Arabic dialects are an analytical
and simplified version of “the synthetic language system which was their
starting point” (Holes, 1995, p. 157). They share many features, including
lexical items, morphological patterns (e.g., verb patterns), negations, and word
order, but differ in others, namely the phonological processes (particularly
between the Western and Eastern dialects) and lexical and idiomatic usage.
These differences evolved from sociological, cultural and historical
circumstances, and they are not systematic (Holes, 1995). Phonological and
lexical differences exist even between regional dialects of the one country, but
are less problematic (Holes, 1995; Watson, 2002). Arab speakers of inter- and
intra-dialectal varieties overcome their linguistic differences by switching to
MSA or to each other’s dialects (cf. Versteegh, 2001).
The range of conversational strategies available for a speaker is socially
determined – “an individual’s set of habitual strategies is unique within that
range” (Tannen, 1982, p. 218) – and the speaker-listener cannot be idealised
as belonging to a homogeneous language community (Foulkes & Docherty,
2006). Moreover, being familiar with a dialect minimises but does not rule out
dialectal problems that might be encountered by EAR translators. Assuming
that the translator is working within his/her range of dialectal expertise, the
universal ethical rules of accuracy for translators and interpreters should apply
when problems of comprehension are encountered due to idiolect (a person’s
individual speech pattern) or communal dialect (a community’s speech pattern
that is geographically, socially, culturally, and/or ethnically determined). This
includes disqualifying oneself from the assignment or, if the problems are
isolated, seeking assistance of colleagues or native speakers who are familiar
with the language variety. With regard to speech production and reception,
studies relating to between-speaker differences (e.g., Smirnova et al., 2007),
between-listener differences (e.g., Grabe et al., 2005) and between-gender
differences (e.g., Rosenhouse, 1998), within the same spoken variety, are
work in progress but worthy of investigation in the context of the present
topic.
Apart from its generic features and language-specific dialectal diversity,
peculiar features of the speech type at hand have a bearing on the
translation/transcription method and process. Listening device recordings are
often of low quality compared with telephone interceptions. Because the
conversation is private in the EAR material and all interlocutors – save any
undercover agent, if involved – are unconscious of the recording of their
conversation, the translator often lacks the necessary contextual knowledge for

Translation  &  Interpreting  Vol  8  No  2  (2016) 48  

interpretive decisions (Bucholtz, 2000). The purpose of the surveillance
exercise is to probe into private life for prosecution, and with this end in view
the police are on the lookout for criminal intention, conspiracy or confessions
of commission of crime. The use of code-switching is a common feature of
EAR, with languages other than English (LOTEs) being strategically mixed
with English (which is often the speakers’ first language) and sometimes Pig
Latin (by juveniles, e.g., abbingstay: stabbing) if the interaction is taking
place in an English-speaking country, to conceal messages or identity.

2.1 Addressing equivalence

To date, discourse-analysis research has rarely concentrated on prosodic
features in interpreting (cf. Merlini & Favaron, 2005; Shlesinger, 1994), and,
to my knowledge, never on the transcription of prosodic features in translated
EAR. As previously discussed, these transcripts are conceived as prose
‘written to be read’ and not ‘written to be read as if spoken’. Spoken and
written languages are packaged differently (cf. the ‘oral-written dichotomy’ of
Horowitz and Samuels (1987, p. 9)), and translating recorded material
imposes the added responsibility of decision-making about the transfer of
meaning using equivalent target language speech conventions (cf. Bucholtz,
2000). This transfer militates against the constraints of the ‘verbatim’
translation requirement, which rules that when ‘interpreting’ EAR – i.e.
transcribing and translating it – “changing or paraphrasing of what is
originally said is the same as altering testimony which is considered perjury or
lying under oath” (Teichman, 2000, para. 3).
The production of an equivalent effect on the users of transcripts can be
achieved through the faithful reconstruction of the interlocutors’ participation
in the formation of the message (cf. Nida, 1964). This requires making use of
conversation analysis with reference to the prosodic and paralinguistic cues of
speech to infer, translate and document meaning.
Selting and Couper-Kuhlen (1996) argue for the implication of prosody
on meaning, and Gumperz points out “the shifts in intonation, volume, rhythm
and tempo, that underlie prosodic assessments, and [explain] their
grammatical functions” (1996, p. xi). He considers that prosody, among other
indexical signs, including code- and style-switching, interacts “with symbolic
(i.e., grammatical and lexical) signs, cultural and other relevant background
knowledge (i.e., contextualization cues) to constitute social action” (as cited in
Prevignano & di Luzio, 2003, p. 9) and mark, analytically, thematic coherence
in various speech events (cf. Antonis et al., 2001; Halliday, 1994; Local,
1996).
Notwithstanding the hypothetical points of departure of the above
interactional, linguistic and sociolinguistic approaches, the concern of this
paper is the consensus on the integral function of prosody in framing meaning
and its universal role in achieving “the pragmatic conditions of
communicative tasks” (Gumperz & Gumperz, 1982, p. 12).
Time pressure, contemporaneousness and live interaction, which are the
three pertinent factors that have driven research into spoken interpreting, are
remotely relevant to the translation of EAR. Also, interpreting skills with
regard to the prosodic features of spoken conversation have been given
marginal attention compared with linguistic features, despite their impact on
the output – be it ‘spoken from spoken’ material (cf. Shlesinger, 1994),
‘spoken from written’ material (cf. Agrifoglio, 2004) or ‘written from spoken’
material.
The genre and type of recorded voice conversation have not been tackled
by translation studies, and the scarce discussions of translation ‘techniques’ of
oral conversation (e.g. Edwards, 1995; Esposito, 2001) or transcription

Translation  &  Interpreting  Vol  8  No  2  (2016) 49  

techniques of monolingual EAR (e.g. Fraser, 2003), have not provided insight
into the transfer of intention and actions, or about other issues, such as
accuracy, clarity and naturalness.
In the absence of research on the topic in translation and interpreting
studies, and given the relevance of prosody to the documentation of oral
conversation, it is worth exploring the extension of application of pertinent
notions in conversation analysis, in particular prosodic and paralinguistic cues,
to the translation of EAR. The hypothesis is then that prosody in conversation
is integral to the translation of EAR and failure to process and document
prosodic features may result in misinterpretation (i.e. misinferencing) and
mistranslation.

2.2 Prosody and meaning in transcripts

Gardner posits that, in conversation analysis, “the transcription process is the
analysis” (1994, p. 103). He argues that, in order to obtain observable data, the
process must not ‘freeze’ the interaction into a text, but rather ‘re-do’ the
event being analysed and ‘create’ “the meanings by the participants from
moment to moment, and document the minute details of the interaction”
(1994, p. 103). Such process revelation is axiomatic given that conversation
involves roles and power relationships, exhibited in structurally organised and
not chaotic exchanges, in spite of the presence of phenomena such as
interruptions and simultaneous talk. In fact “nothing in conversation can be
dismissed as disorderly, accidental [or] irrelevant” (Gardner, 1994, p. 102).
Merlini and Favaron (2005) conducted an insightful investigation on power
relationships and voice of interpreting in speech pathology, using prosody and
conversation analysis. Their analysis shows the intrusive role that can be
played by the interpreter who prosodically mismanages and manipulates the
traditional doctor-patient interaction.
Examples of paralinguistic cues include ums, mms, uhs or uh-huhs in
English, which can express hesitation, incomprehension and show that the
interlocutors are listening or pretend to be listening to each other (cf. Sacks,
1995, pp. 746-747). Apart from its use as a way of ‘sudden remembering’
(Jefferson, as cited in Local, 1996, p. 178), oh can indicate a ‘change-of-state
token’ (Local, 1996, p. 178) when its phonetic parameters are analysed with
reference to lexis and syntax, e.g., displaying ‘news-receipt’ or ‘partial repeats
of prior turn’ (Local, 1996, pp. 180-201). Other ‘intervals and between
utterances’ (Gardner, 1994, pp. 185-186), i.e., pauses, and prosodic features,
include coughing, laughter (indicating for example intimacy), and sound
prolongation. These cues are relevant in the translation of EAR material,
which also includes the important factor of inter-language transfer. A brief
pause, for example, can indicate “a ‘slot’ which the referring speaker leaves
open for the recipient to insert, in the case of success, a token of recognition”
(Müller, 1996, p. 135). In this sense, translating EAR is largely influenced by
the chosen transcription method.
The translator of other types of audio recordings (e.g. focus group) may
have a licence to deliver pragmatic meanings in the prose, in parenthetical
comments (e.g. “expressing anger”) or in footnotes, without recourse to the
transcription of prosodic features. This is a luxury the translator of EAR does
not, and should not, have. It is noteworthy that this paper considers prosody as
a universal property of language and is solely concerned with its interactional
use. This includes the employment of the same prosodic feature for different
functions, e.g., tune (O’Connor & Arnold, 1973), pause, latching, and
different prosodic features for similar functions (e.g., the feedback tokens
yeah and okay, sound prolongation).

Translation  &  Interpreting  Vol  8  No  2  (2016) 50  

2.2.2 Function of intonation. The prominence of prosodic features, such as
intonation, to meaning is not new. As succinctly put by Shlesinger (1994, p.
225), “The functionality of intonational choices and their role in facilitating
(or obstructing) communication is by now a universal point of departure in the
literature.”
Antonis et al. (2001) identify three notions around which the ‘structural’
functions of intonation are centred: prominence (provision of weight
structuring of linguistic units, i.e., stress distinction on syllables), grouping
(provision of coherence and segmentation of linguistic units into prosodic
units, i.e., provision of tonal prominence or focus) and discourse (structuring
prosodic units according to topics of discussion and turn regulations between
speakers in a conversation). Halliday (1994) posits three intonation functions
in spoken language: grammatical (relating to mood structure, e.g.,
question/statement, and identification of clause and sentence constituents),
attitudinal (relating to speakers’ attitudes and emotions, e.g., surprise,
indifference, irony, gratefulness) and informational (indicating importance of
a word and marking given and new information).
Couper-Khulen and Selting’s (1996) interactional approach converges
with the preceding functions of intonation outlined by Halliday (1994) and
Antonis et al. (2001), but extends beyond these functions to embrace prosody
in general. Crystal (1969) also argues that the meaning of intonation works in
tandem with other variables, including, prosodic and paralinguistic systems,
lexis, style, and kinetic cues. Prominence, grammatical, attitudinal, and
discursive objectives inform then prosodical choices in speech. As they are
conventional and acquired (cf. Cook, 2002), the tools with which these
objectives are achieved are largely finite.
Bolinger also posits that intonation is located within the general scheme
of iconic nonverbal communication and defines intonation as “primarily a
symptom of how we feel about what we say, or how we feel when we say”
(1989, p. 1). In a chapter dedicated to ‘dialect and language’, Bolinger (1989,
pp. 26-64) argues for the universality of prosody, and in particular intonation
across a variety of languages, including Syrian Arabic (cf. Syro-Lebanese
dialectal classification in Versteegh, 2001). With reference to Arabic and
English, Bolinger concludes that “the similarity between the two languages is
remarkable, including certain rather small details” (1989, p. 54). Intonation is
suprasegmental, and despite their segmental differences, Arabic and English
are phonologically stress-timed. Moreover, like English, Arabic can take no
more than one stressed syllable per word (cf. Holes, 1995, on stress in Modern
Standard Arabic and Ghazali et al., 2002, on stress-timing variation in Arabic
dialects). Apart from the consensus on the difference in intonation in tag and
wh-questions between English and Arabic (cf. de Jong & Zawaydeh, 1999;
Odisho, 2005), significant prosodic similarities between English and Arabic
are reported by researchers other than Bolinger (e.g. Chahal, 1999; de Jong &
Zawaydeh, 1999; Holes, 1995; Jun, 2005; Odisho, 2005).
The phonological, prosodic and intonational closeness between English
and Syrian Arabic (the dialect of the speech excerpt in this paper) provides the
ground for the use of approaches to the analysis of interactional and functional
aspects of prosody and intonation described in the literature on the English
language for inference, translation and transcription of EAR material. This is
supported by the fact that the focus of this paper is on the emotions provided
by prosodic features, rather than on the speaker and word recognition, accent
identification, and topic and sentence segmentation. Furthermore, the
processing of segmental information in speech is closely linked to stress.
Languages and dialects, as commonly held, differ markedly in the details
of phonetic realisation of prosodic patterns (see Gibbon, 1998; Kulk et al.,

Translation  &  Interpreting  Vol  8  No  2  (2016) 51  

2003), but influential research on intonation functions and patterns (see
Bolinger, 1978, 1989; Lieberman, 1967; Pierrehumbert & Hirschberg, 1990)
is based on a universalist approach to intonation rather than a typological one
(see discussion on both approaches in Ladd, 2001), or an approach which has
gained universality through application (see Fujisaki, 1983; O’Connor &
Arnold, 1973). O’Connor and Arnold (1973) made a pioneering study on the
role of intonation in emotional expressions in colloquial English (middle-class
southern British English). Their seminal work remains an emblematic resource
for learning intonation in English teaching and in research on intonation (cf.
Grabe et al., 2005; Gussenhoven, 2004). O’Connor and Arnold (1973) identify
seven tunes, or pitch treatments, that are used to create intonation for syllables
and words, and hence signal intentions and attitudes and provide a means of
steering inferential processes.
They also systematically identify ten tone groups in spoken English, each
of which expresses different sets of attitudes in different contexts and its tunes
have one or more pitch feature in common. These tone groups are: low drop
(expressing, for example, reservedness in statements or intensity in wh-
questions); high drop (e.g. demanding agreement with question tags); take-off
(e.g. appealing to the listener with commands); low bounce (e.g., encouraging
effect with interjections); switchback (e.g. astonishment in echoed questions);
long jump (e.g., protestation in statements); high bounce (e.g., tentativeness in
straightforward wh-questions); jack-knife (e.g., antagonism with yes-no
questions); high dive (e.g. pleading with commands); and terrace (e.g. calling
out to someone with interjections).
O’Connor and Arnold’s (1973) categorisation of ‘tone groups’ is
appropriate for the analysis of attitudes and emotions conveyed by speakers in
evidentiary audio recordings, as it is in tune with the interactional, linguistic
and sociolinguistic approaches to the function of intonation as discussed
above. The relevance of the authors’ categorisation is also prompted by the
speech pair used in the excerpt, which is the Syro-Lebanese variety of Arabic
and Australian English, described by Gupta (1997) as ‘monolingual ancestral
English’.

3. Procedure and instrument

To test the above assumptions, an authentic Arabic speech excerpt from an

evidentiary audio recording was faithfully transcribed, translated and
analysed. The excerpt is from a speech of one hour and ten minutes, recorded
by a crime squad in an Australian gaol using a listening device. The
conversation is in Lebanese Arabic mixed with English, between two males
charged with a criminal offence and awaiting a court hearing. The main
criterion for choosing a conversation recorded through a listening device is to
highlight the potential unintelligibility and incoherence that confront the
transcriber, and discuss the relevance or otherwise of incorporating prosodic
and paralinguistic information among the features of intelligible speech.
In discussing the meaning of representative examples, consideration will
be given to interlocutors’ style, monitoring, planning, and control in the
conversation, with reference to their speech presentation on transcripts.

3.1 Transcription conventions

Crime agencies have opted for a minimalist approach to the documentation of
prosodic features – restricting them, often cosmetically, to feedback, receipt
tokens and hesitation (see Table 1 below). The transcription conventions need

Translation  &  Interpreting  Vol  8  No  2  (2016) 52  

to serve as an analytical instrument of the translated material and suit the
purpose of producing a legible, but also defensible, transcript.
The transcripts of EAR are to be read by non-linguists and non-
conversationalists, including judges, lawyers, police investigators, jury
members, and the accused. According to Duranti (1997, p. 142) “the process
of transcribing implies a process of socialization of our readers to particular
transcribing needs and conventions”. For the purposes of the present work, the
transcribed translations need to be faithful to the original, read as fluently as
possible, and at the same time be highly informative as to the prosodic and
paralinguistic cues of the spoken utterances.
In conversation analysis, different scholars have used different transcription
conventions to serve their research needs. Paul ten Have (1999) discusses the
generic information available in monolingual transcripts, including archival
information, such as time, date, and inferential information, such as
overlapped speech and stresses. He and others (e.g., Edwards, 1995; Fraser,
1996) warn against the contentious issue of voice identification. The IPA
transcription format is largely inaccessible, although it may have its use in
computational and phonetic analysis. Substituting general rules of orthography
to indicate phonetic features (e.g. ‘dz’ for ‘does’ and the vernacular ‘yer’ for
‘your’) is also, obviously, impractical and does not serve the purpose (Duranti,
1997). Accordingly, a hybrid transcription system, selected from ten Have
(1999), Gumperz and Roberts (1991) and Gardner (1994), is adopted in the
translation of the excerpts below. The selection aims at covering prosodic and
paralinguistic cues pertinent to deciphering meaning, minimising visual
disruption, and providing relative ease in word-processing.

Table 1. Transcription Conventions

Symbol Significance
Simultaneous, overlapping and latched utterances
[] In overlap, the point of onset is marked with left hand square
brackets, and the point at which overlap stops is marked with
right-hand square brackets.
= Equal signs link contiguous stretches of talk between which
there is no gap and no overlap (latched).
= and [ ] If more than one speaker latches onto a previous turn, this is
shown through a combination of equal sign and square bracket.
Prosodic features of utterances
(0.0) Extended pause- Numbers in parentheses indicate elapsed time
in silence by tenth of seconds, so (7.1) is a pause of 7 seconds
and one tenth of a second.

Silences can occur within utterances (pauses), or between

utterances (gaps).
(.) Brief pause - A dot in parentheses indicates a pause of less than
0.2 seconds within or between utterances.
word Underscoring indicates some form of stress, via pitch and/or
amplitude.
:: Colons indicate prolongation of the immediately prior sound.
Multiple colons indicate a more prolonged sound.
- A dash indicates a sudden cut-off (truncation) of the current
sound.
.,? Punctuation marks are used to indicate characteristics of speech
production, especially intonation; they do not refer to
grammatical units.
. A full stop indicates a falling terminal contour, i.e., a ‘final’
intonation.
, A comma indicates a continuing intonation (contour) fall in tone,
like when reading items from a list. It represents talk that is
hearably incomplete.
? A question mark indicates a strongly rising intonation (contour).

Translation  &  Interpreting  Vol  8  No  2  (2016) 53  

 Its characterising feature is that it rises a long way in pitch and
ends up at the high end of the pitch range.
The absence of an utterance-final marker indicates some sort of
‘indeterminate’ contour.
↑$ Arrows indicate marked shifts into higher or lower pitch in the
utterance part immediately following the arrow.
WORD Upper case indicates especially loud sounds relative to the
surrounding talk.
◦ Utterances or utterance parts bracketed by degree signs are
relatively quieter than the surrounding talk.
>words< Right/left pointing carets bracketing an utterance or utterance-
part indicate speeding up.
<words> Left/right pointing carets bracketing an utterance or utterance-
part indicate slow talk. It is optional as it can be apparent from
colons indicating lengthening.
Feedback, receipt tokens and hesitation
Continuers and completers: okay, uh hm, mm hm, mm, oh,
yeah, right, alright, you know
Hesitation marker: uh, uhhh, for short or longer hesitation,
respectively.
Translator’s doubts and comments
() Empty parentheses indicate inability to hear what was said. The
length of the parenthesised space indicates the length of the
untranscribed talk.

In the speaker designated column, the empty parentheses

indicate inability to identify the speaker.
(( )) Double parentheses contain transcriber’s/translator’s description
rather, or in addition to, transcriptions, i.e., vocalisations or non-
linguistic vocal effects that indicate voice qualifications but
cannot be satisfactorily transcribed in symbols.

3.2 Translation method

The constraints imposed on the translator by the outlined speech-type purpose
and by the obligation of producing ‘verbatim’ transcripts imposed by crime
agencies and the legal system, clearly demonstrate the appropriateness of
Newmark’s (1988) ‘faithful translation’ method for EAR material, as opposed
to the word-for-word or literal methods (1988). Newmark defines faithful
translation as a method that

…attempts to reproduce the precise contextual meaning of the original within the
constraints of the TL grammatical structures. It ‘transfers’ cultural words and
preserves the degree of grammatical and lexical ‘abnormality’ (deviation from
SL norms) in the translation. It attempts to be completely faithful to the
intentions and the text-realisation of the SL writer (1988, p. 46).

Newmark refers here to the translation of texts, which are syntactically

well organised and evenly punctuated. Natural speech, however, exhibits
paratactic constructions (predominant use of coordination and juxtaposition)
instead of the hypotactic constructions in writing (predominant use of
subordination).
In An Introduction to Functional Grammar, Halliday identifies two
different kinds of complexity associated with written and spoken languages as
follows:

Typically, written language becomes complex by being lexically dense: it packs

a large number of lexical items into each clause; whereas spoken language
becomes complex by being grammatically intricate: it builds up elaborate clause
complexes out of parataxis and hypotaxis (1994, p. 350).

Translation  &  Interpreting  Vol  8  No  2  (2016) 54  

 Halliday defines parataxis as “the relation between two like elements of
equal status, one initiating and the other continuing, whilst hypotaxis is the
relation between a dependent element and its dominant, the element on which
it is dependent” (1994, p. 218). From the perspective of syntax, Newmark’s
definition of faithful translation is appropriate, but from the perspective of
punctuation, it is inadequate, as will be discussed in the subsequent sections.
In speech, pauses and stops are communicated phonologically and not
graphically. Interestingly, Halliday (1987, p. 66) argues that spoken discourse
is grammatically more intricate than written discourse, adding that “The more
natural, un-self-monitored the discourse, the more intricate the grammatical
patterns that can be woven.” Earlier, Halliday and Hasan (1976, p. 5) state that
“meaning is put into wording [i.e., words, grammar and syntax], and wording
into sound or writing” (emphasis added). This aligns with the established fact
that prosodic and paralinguistic features can form an integral part of the
meanings of lexico-grammatical structures in the spoken mode.
Drawing on Halliday's work, Eggins (2004, p. 255) argues that when we
talk we often chain clauses in sequence “and use markers to show the
relationship between clauses (e.g., when, because), and we also use the spoken
language systems of rhythm and intonation to signal to our listeners when
we’ve reached the end of a clause sequence” or a clause complex. In systemic
functional linguistics, a clause complex refers to the “grammatical and
semantic unit formed when two or more clauses are linked together in certain
systematic and meaningful ways” (Eggins, 2004, p. 255). Castello (2008)
further argues that lexico-grammatical intricacy is reflected in how many
clauses join together to form a clause complex.
It is this grammatical intricacy and complexity of spoken language which
motivates and informs the translation method to be adopted for EAR material,
with the aim to maintain the stylistic accuracy in line with the rules applicable
to the treatment of register and stylistic variations in interpreting (Hale, 1997,
2002). One handy and accessible feature to account for style is the apostrophe
(’) to signal a missing sound expected in Standard English (cf. Labov as cited
in Duranti, 1997), e.g. ’cause you ain’t goin’ to, rather than the formal
because you are not going to. Intercultural variability requiring the translator’s
intervention to achieve target language equivalence is linguistic and stylistic
as discussed and exemplified.

4. Sample analysis

In the excerpt below, the interlocutors speak the North Lebanese dialect. The
speech is significantly of poor sound quality due largely to noise, background
conversation, echo, and, presumably, the unconsciousness of the interlocutors
of the position of the listening device.
The conversation is mainly in Arabic, exhibiting now and then code-
switching, borrowing of English words and phrases which are adapted to the
grammatical rules of Arabic (Arabised). The social motivation for switching at
the phrase level appears to be for convenience and fluency, e.g., ‫ﻳﯾﻮ ﺭرﻛﻦ ﻓﻴﯿﻮﻥن‬
‫ﻳﯾﺸﻮﻓﻮ‬: ‘you reckon they can see’, while need seems to be the driver behind the
lexical switching, such as ‫ﺷ ّﺮﺝج‬:‘to charge’.
In the following analysis, use is optimally made of the texture, structure
and prosodic features at hand, and reference is made to examples that
highlight the importance of the transcription of prosody onto a stylistically
faithful translation.

Translation  &  Interpreting  Vol  8  No  2  (2016) 55  

Table 2. Excerpt of translation of evidentiary audio recording
MVI: male voice one; MV2: male voice Texts in bold are spoken in English
two.
1 (Noise and echo) (Noise and echo)
2 MV1: ( ) ( )
3 MV2: ( ) ↑did they charge( )? ‫) ( ↑ ﺷﺭرّ ﺟـﻭو) (؟‬
4 MV1: ↑no (1.0) but he <showed( )>. .<( )‫( ﺑﺱس >ﺃأﺭرﺟﺎ‬1.0 ) ‫↑ﻷ‬
5 MV2: they’re ↑goin’ to charge ( ). ( ) ( ) .( )‫↑ﺣﻳﯾـﺷﺭرّ ﺟـﻭو‬
(.) they’re ↑goin’ >to charge ( )<. .>( )‫( ↑ﺭرﺡح <ﻳﯾـﺷﺭرّ ﺟـﻭو‬.)
6 (whistle noise) (3.0) (whistle noise) (3.0)
7 MV1: ( ) ( )
8 ( ) ( )
9 MV2: ( ) ( )
10 MV1: sixty ↑six ‫↑ﺳ ّﺗﺎﻭوﺳﺗﻳﯾﻥن‬
11 MV2: ↑are they goin’ to charge ( )? ‫ﺭرﺡح ﻳﯾـﺷﺭرّ ﺟـﻭو) (؟‬
12 MV1: huh? ( ) they showed me ،٬‫( ﺍاﻟﺻﻭوﺭر‬2.2 ) .‫ﻫﮬﮪھﻪﮫ؟ ) ( ﻓﺭرﺟﻭوﻧﻲ ﻛﻠﺷﻲ‬
everything. (2.2) the photos,
13 MV2: yeah. .‫ﺇإﻳﯾﻪﮫ‬
14 MV1: (1.0) our face. ّ ‫( ﻭو‬1.0 )
.‫ﺷﻧﺎ‬
15 ( ): (◦ ◦) (◦ ◦)
16 MV1: ( ) our face, ( ). .( ) ،٬‫) ( ﻭوﺷ ّﻧﺎ‬
17 (Unintelligible background (Unintelligible background
conversation) conversation)
18 (2.5) (2.5)
19 MV1: but uh ( ) (1.5) in the truck. () (1.0 ) .‫( ﺑﺎﻟـﺗﺭرﺍاﻙك‬1.5) ( ) ‫ﺑﺱس ﺃأﻩه‬
(1.0) ( )
20 MV2: ( )= =( )
21 MV1: =the three jobs I did in the car. .‫=ﺍاﻟﺗﻼﺕت ﺟﻭوﺑـﺍاﺕت ﻳﯾﻠﻠﻲ ﻋﻣﻠﺗﻥن ﺑﺎﻟﺳﻳﯾﺎﺭرﺓة‬
22 ( ) (1.0) (1.0 ) ( )
23 MV1: the ↑photos. ( )(.) they’ve talked (.) .( ) ‫( ﺣﻛﻳﯾﻭو ﻣﻊ‬.) ( ) .‫ﺍاﻟـ↑ﺻﻭوﺭر‬
with( ). (.)
24 MV2: $yeah. .‫ﺇإﻳﯾﻪﮫ‬$
25 MV1: they’ve talked with him. .‫ﺣﺎﻛﻳﯾﻳﯾﻥن ﻣﻌﻭو‬
26 MV2: ↑did he talk with( )? ‫(؟‬ )‫↑ﺣﺎﻛﻳﯾﻳﯾﻥن ﻣﻊ‬
27 MV1: ↑yeah.= =.‫↑ﺇإﻳﯾﻪﮫ‬
28 MV2: =↑what did he say? ‫= ↑ﺷﻭو ﻗﺎﻝل؟‬
29 MV1: (.) I don’t know. .‫( ﺃأﻱي ﺩدﻭوﻧﺕت ﻧﻭو‬.)
30 MV2: ( ) ( )
31 (Noise, echo and unintelligible (Noise, echo and unintelligible
background conversation) background conversation)
32 (2.0) (2.0)
33 MV1: the photos of the house, ( ) ( ) ،٬‫ﺻﻭوﺭر ﺍاﻟﺑﻳﯾﺕت‬
34 (2.8) (2.8)
35 ( ) ( )
36 (unintelligible background (unintelligible background
conversation) conversation)
37 MV2: ( ) $you reckon >they can see ‫ ﻳﯾﻭو ﺭرﻛﻥن <ﻓﻳﯾﻭوﻥن ﻳﯾﺷﻭوﻓﻭو) (>؟‬$ ( )
( )<?
38 ( ) ()
39 MV1: ↑our face? ّ ‫↑ﻭو‬
‫ﺷﻧﺎ؟‬
40 (unintelligible background (unintelligible background
conversation) conversation)
41 (2.5) (2.5)
42 MV1: all of them. ↑all of them. .‫ ↑ ﺃأﻭوﻝل ﺃأﻭوﻓﻡم‬.‫ﺃأﻭوﻝل ﺃأﻭوﻓﻡم‬
43 MV2: ↑huh? ‫↑ﻫﮬﮪھﺎﻩه؟‬
44 (2.0) (2.0)
45 ( ): ( ) there are knuckledusters. .‫) ( ﻓﻲ ﺑﻭوﻧﻳﯾﺎﺕت‬
46 (unintelligible background (unintelligible background
conversation) conversation)
47 (1.5) (1.5)
48 MV2: ( ) ( )
49 MV2: there’s the ↑shirt. Inside, it’s (4.2) .‫ ↑ﻣﺑﻳﯾّﻧﻲ‬،٬‫ ﺟﻭوﺍا‬.‫ﻓﻲ ↑ﺍاﻟﻘﻳﯾﻣﺹص‬
↑clear. (4.2)
50 (unintelligible background (unintelligible background
conversation) conversation)

Translation  &  Interpreting  Vol  8  No  2  (2016) 56  

 51 (1.5) (1.5)
52 (unintelligible background (unintelligible background
conversation) conversation)
53 MV1: but (4.0) in the photos it’s ( ), (.) (.) ،٬( ) ‫ﺑﺱس ) ( ﺑﺎﻟﺻﻭوﺭرﻫﮬﮪھﻲ‬
54 other than the shirt, ( ) the shirt, ( ) ‫( < ↑ﻳﯾﻭو ﻧﻭو ﻭوﺍاﺕت ﺃأﻱي‬ ) ،٬‫ ) ( ﺍاﻟﻘﻣﻳﯾﺹص‬،٬‫ﻏﻳﯾﺭر ﺍاﻟﻘﻣﻳﯾﺹص‬
>↑you know what I mean?< >‫ﻣﻳﯾﻥن؟‬
55 (unintelligible background (unintelligible background
conversation) conversation)
56 ( ) ( )
57 MV1:() there nothing except the ‫( َﻭوﻥن ﻫﮬﮪھﻧﺩدﺭرﺩد‬.) ( ) ‫ ﻓﺭرْ ﺟﺎﻧﻲ‬.‫ﺻﻭوﺭر‬$ ‫)( ﻣﺎ ﻓﻲ ﺇإﻻ ﺍاﻟـ‬
$photos. he showed ( ) to me (.) one .‫ﺑﺭرﺳﻧﺕت‬
hundred per cent.
58 ( ) (background noise and echo) (background noise and echo) ( )
59 MV2: They’re goin’ to charge you, I .‫ﺭرﺡح ﻳﯾـﺷﺭرّ ﺟـﻭوﻙك ﻗﺎﻝل‬
heard.
60 MV1: ( )? ‫(؟‬ )
61 MV2: they’re goin’ to charge you (.) ( ) ( ) (.) ‫ﺭرﺡح ﻳﯾـﺷﺭرّ ﺟـﻭوﻙك‬
62 (2.5) (2.5)
63 MV1: thirty one ( ). .( ) ‫ﻭوﺍاﺣﺩد ﻭوﺛﻼﺛﻳﯾﻥن‬
64 MV2: ↑thirty one? (3.0) ↑serious. .‫( ↑ﻻ‬3.0 ) ‫ﻭوﺍاﺣﺩد ﻭوﺛﻼﺛﻳﯾﻥن؟‬
65 MV1: ( ), ↑thirty one. .‫ ↑ﻭوﺍاﺣﺩد ﻭوﺛﻼﺛﻳﯾﻥن‬،٬( )
66 MV2: ↑they’re goin’ to charge you ‫ﺭرﺡح ↑ﻳﯾﺷﺭرّ ﺟﻭوﻙك ﻓﻳﯾﻭوﻥن ↑ﻫﮬﮪھﻠﻕق؟‬
with them ↑now?
67 MV1: I don’t know, ( ) ( ) ،٬‫ﺃأﻱي ﺩدﻭو ﻧﺕت ﻧﻭو‬
68 MV2: (.) they showed ( ), but they ( )‫ ﺑﺱس ﻣﺎ ﻗﺎﻟﻭو‬،٬( )‫( ﻓﺭرﺟﻭو‬.)
didn’t tell ( )
69 ( ) (background noise and echo) (background noise and echo)( )
70 MV1: ( ) they’re trying ( ) to make .‫) ( ﻋﻡم ﻳﯾﺟﺭرﺑﻭو ) ( ﺣﺗﻰ ﻳﯾﺧﻠّﻭوﻧﻲ ↑ﺇإﺣﻛﻲ‬
me ↑talk.
71 ( ) (background noise and echo) (background noise and echo) ( )
72 MV1: ↑you know what I mean? ( ) ( ) ‫↑ﻳﯾﻭو ﻧﻭو ﻭوﺍاﺕت ﺃأﻱي ﻣﻳﯾﻥن؟‬
73 ( ) (background noise and echo) (background noise and echo) ( )

The meaning of the utterance in row 4 and the first utterance in row 5 is
determined by the low drop tone group in both, which conveys a definite and
complete opinion, marked by the absence of a nuclear head (i.e., the first
stressed word or syllable in the tone group), hence expresses detachment
(O’Connor & Arnold, 1973, pp. 47-48). By contrast, the second utterance in
row 5 ‘they are ↑goin’ >to charge ( )<’ has two different features: (1) its tone
group is ‘high dive’ (O’Connor & Arnold, 1973, pp. 82-88), which also
expresses unreservedness, emphatic definiteness and completeness through the
accented (stressed) ‘charge’, and (2) the accelerating tempo exhibited
through the ‘clipped syllable’ (word) ‘charge’. This corresponds with
Crystal’s findings that ‘clipped syllables’ frequently co-occur with a nuclear
tone (here, ‘charge’), “regardless of other pitch features co-occurring, which
suggests that this feature is an important modifying factor in the interpretation
of pitch glides” (1969, p. 154) . Further, based on an experiment involving the
application of various feelings and intentions to statements by a number of
subjects, Crystal (1969, p. 305) found that fast tempo is used to express
conspiracy, among other uses, which is relevant to the topic at hand. This
gives rise to the contrast in meaning, between the two adjacent and identical
utterances, i.e., ‘they will charge someone?’ in the latter utterance as opposed
to the former ‘they may charge someone’. This suggests that the
documentation of accent, intonation and tempo could be crucial.
Noteworthy also is the unintelligibility of the personal pronoun affixed
to the verb endings in rows 3, 4 and 5, which in Arabic refers, grammatically,
to the object. In rows 3 and 5, MV2 says, ‫) ( ﺷ ّﺮﺟﻮ‬: ‘they charged ( )’. Two
possible objects can be inferred: ‘you’ or ‘him’. The missing part of the last
syllable ‘you’ ‫ = ﺷﺮﺟﻮﻙك‬charra/juk: ‘they charged you’, or ‘him’ ‫= ﺷ ّﺮﺟﻮﻩه‬
charra/juh: ‘they charged him’, excludes the object ‘me’, given that no other

Translation  &  Interpreting  Vol  8  No  2  (2016) 57  

syllable is heard (compare ‫ = ﺷﺮﺟﻮﻧﻲ‬charra/ju/ni: ‘they charged me’). The
verb ‘charraja’ is a loan word from English. The last possibility is also
inconsistent with the context and legal process, knowing that a person is told
when they are charged. Guesswork can be detrimental here, so the use of ( ),
i.e., unintelligible, is important, rather than the arbitrary translator’s/
transcriber’s comment of ‘sounds like’.
In row 4, the co-text (the adjacency pair, question/answer) combined with
the decelerated tempo of ( ) ‫ﺃأﺭرﺟﺎ‬: he showed ( ), suggest two possible
‘objects’: ‘me’ or ‘him’.
In row 15, the degree marks in the unintelligible brackets imply whisper,
the meaning of which can be important in other contexts as it may evoke
conspiratorial intention.
In row 24, the falling tone in ‘yeah’ communicates an invitation to the
speaker to ‘continue’ as in mm hm.
The latching in rows 27 and 28 expresses MV2’s concern in this context,
but could equally express power in a larger context.
The pronoun in row 37 is not clear; it sounds like ‘us’ or ‘me’. However,
the following clarifying question, ‘our face?’ indicates the likelihood of ‘us’.
The accelerated tempo in row 37 conveys impatience and irritability (cf.
Crystal, 1969, p. 305) about the fact that ‘they can be identified from the
photos’.
In row 39, the noun-pronoun ‘our face’ disagreement in number is
common in spoken Arabic and exhibits, through the faithful translation
method adopted, informal colloquial usage but not the speaker’s educational
level in Arabic.
The 2.5 second pause in row 41 helps in delivering the message, given
that the gap between the clarifying rhetorical question in row 39 (also
expressed through intonation), ‘do you mean our faces?/you mean our faces?’
and the second heard utterance by the same speaker (row 42) is relatively
short and interrupted by something said. Hence, the utterances are clearly an
emphatic answer to the question asked originally by MV2 in row 37.
In row 59, ‫ ﺭرﺡح ﻳﯾﺸﺮﺟﻮﻙك ﻗﺎﻝل‬literally means ‘they’re going to charge you, he
said’; however, in the absence of a referent, the end position of ‘he said’ in
this instance, imparts ‘I heard’. This is also a topic shift/opener strategy in
Levantine (also known as Greater Syria, including, linguistically, Lebanon)
Arabic conversation, expressing knowledge that the listener already has.
The low drop tone group in row 61 conveys a detached statement and
not a question. The above discussion demonstrates that prosodic features can
be very crucial in certain instances, to infer meaning despite the absence of
linguistic information.
In sum, the faithful translation method applied to the excerpt above
would have generated serious comprehension problems to the reader/analyst
without recourse to prosodic features. To put this ‘uncompromising and
dogmatic’ translation method on a par with the ‘flexible’ semantic translation
(Newmark, 1988, p. 46), use was made of available prosodic features, among
other strategies. Therefore, Newmark’s (1988, p. 46) definition of faithful
translation of EAR calls for the following extension:
Translating evidentiary audio recordings must be conducted through a
faithful translation method that attempts to reproduce the contextual meaning
of the original speech within the constraints of the TL grammatical structures.
It transfers cultural words and preserves the degree of grammatical and lexical
abnormality (deviation from SL norms). It attempts to be completely faithful
to the intentions and the speech-realisation of the SL speaker via linguistic,
paralinguistic and prosodic means as necessary.

Translation  &  Interpreting  Vol  8  No  2  (2016) 58  

 Table 3 below illustrates some significant strategies used to optimally
manage each component of the faithful translation rationale. The right hand
column shows possible erroneous interpretation in the absence of prosodic
and/or paralinguistic cues.

Table 3: Intentions, attitudes or feelings marked by prosodic and paralinguistic

cues
Speech strategy of Intention marked Possible erroneous
prosodic or by prosodic and interpretation without added
Example
paralinguistic paralinguistic prosodic and paralinguistic
feature/s feature/s feature/s
They’re goin’ to charge ( ).

they’re goin’ >to charge - with a jack-knife intonation (rise-fall

High dive intonation, Unreserved and
( )<. tune ending):
Fast tempo and emphatic definite
Disclaiming responsibility
accentuation statement.
(Row 5)
- with a switchback intonation (fall-
rise tune ending): Concerned.
“they’re goin’ to charge you, he
A falling terminal tone
They’re goin’ to charge A topic shift/opener said.”
group on tag
you, I heard. device inviting the
statement (full stop)
listener to chat about - with a switchback intonation (fall-
with absence of
(Row 59) shared information. rise on tag statement):
referent.
Questioning with a tone of surprise.
Managing intentions marked with non-intonational devices
Speech strategy/
Contextual
Example turn taking/doubts Possible literal meaning
meaning
/comments
no (1.0) but he Marking unintelligible
<showed ( ) Unintelligible syllable inflected pronoun. In
No one-to-one correspondence.
(unintelligible syllable)>. (as opposed to word) the context, either ‘me’
(Row 4) or ‘him’.
MV1: yeah.=
MV2: = what did he say? Latching Conveying concern. Exhibiting power.
(Rows 27-28)

Table 3 provides a snapshot of intentions ‘concluded’ with reference to

prosodic features. These cues, in tandem with the lexico-grammatical
organisation, influence, and are influenced by the context and moods at the
utterance, sequence and speech levels. Although a limited number of
utterances are analysed, the discussion provides plausible hints about the
overall mood of the excerpt: the interlocutors’ concern about their situation
and admission of involvement in a joint criminal enterprise.
The analysis has identified stretches of texts to which the prosodic
analysis has no bearing, which serves to minimise distraction in the
transcription and reading processes. Although the transcription conventions
are applied indiscriminately in documenting crucial intended messages, they
can be selectively used where they have significant relevance to meaning.
According to ten Have (1999, p. 78), transcriptions “are always and
necessarily selective.” Analysts have traditionally devised interactional
features transcribed to document details to texts written in standard
orthography, according to their interest and need. The same principle applies
to the transcription of EAR.

5. Putting theory into practice

The above discussion suggests that EAR translators must have the necessary
knowledge and knowhow to account for the linguistic and stylistic features of
their working dialects, and be aware of the interactional role of prosody (and
Translation  &  Interpreting  Vol  8  No  2  (2016) 59  
its transcription) in these dialects. Only trained translators who are native
speakers of the source language dialect can meet these criteria.
The discussion also suggests that analysts in crime agencies in charge of
analysing EAR translations for prosecution purposes need to have specialised
training in conversation analysis. It is equally desirable for legal professionals
to become familiar with the rationale and significance of prosody in
transcripts and of the outlined translation method.
Translation and interpreting training programs, especially those
encompassing legal interpreting and specialised translation, ought to include
training in decoding and encoding prosody and paralinguistic cues in speech.
The theory and practice of conversation analysis should be integral to
interpreting courses at all levels, and to translation courses at an advanced
level. In-house workshops are the obvious training forum for translators
practising in the field and clients interested in EAR translations, particularly
police investigators and analysts.

6. Conclusion

Many EAR translations become crucial pieces of evidence in courts. The

success of pleadings and the futures of those under investigation can hinge on
the precision of the translation and transcription. These tasks are usually
performed from recorded audio conversations that lack visual aids such as
non-verbal expression (looks or gestures) and the benefit of feedback that
face-to-face interpreters enjoy. The occasional poor sound quality, the privacy
of conversation, the regional and idiosyncratic dialectal problems, and the
restrictions imposed on the provision of ‘processed’ translation, all lead to the
production of texts which are tantamount to unintelligible words and
sentences. The adoption of a minimalist approach to account for non-linguistic
cues in the translation of conversations does not serve the purpose. Face-to-
face interpreters (should) rely on, and (should) relay prosodic features, in
particular intonation, to determine and transmit intentions, attitudes and
feelings. The same process should apply to the translation of EAR. The
literature on conversation analysis, prosody and intonation provides insights
into the structure and realisation of meaning of the spoken language, the
material in question. These studies have developed transcription conventions
that can be employed to complete the puzzle of the simplistic and risky
approach of ‘faithfully’ translating and transcribing speeches through standard
orthography only. Here an evidentiary audio recording excerpt was
comprehensively transcribed and faithfully translated and using conventional
transcription symbols. The analysis proves that a number of these symbols are
essential for the reader to infer the full meaning and mood of utterances, turn-
takings, and ultimately the speech itself, and hence their incorporation in the
transcript is necessary.
In the context of EAR translation, the interpretation, or reception, of
prosodic features reproduced in English – the target language in this paper -
rests with the crime agency’s analysts, who must also match the translator’s
expertise in terms of use and usage of prosody and conversation analysis.
Pedagogically, the appreciation of conversation strategies and prosodic
features, in addition to the framing and delivery of meaning, can help enhance
the cognitive competence required by trainee interpreters for the processing
and storage of information, and also assist in note-making and delivery.
Although the above claims hinge on the analysis of a short sample with
reference to one speech pair, it is assumed that the translational,
transcriptional and analytical methodology and instrument used could yield

Translation  &  Interpreting  Vol  8  No  2  (2016) 60  

similar results if applied to other speech pairs. Further investigation into the
translation and transcription of conversation between interlocutors speaking
different dialects or interlocutors whose dialects are different to that of the
translator should provide overdue insight to those who are directly or
indirectly concerned. Also, further empirical examination of the impact of the
approach adopted here on the users of transcripts, including the translator as
an ‘expert earwitness’, would demonstrate the relevance of the procedure to
the justice system.

References

Agrifoglio, M. (2004). Sight translation and interpreting: a comparative analysis of

constraints and failures. Interpreting, 6(1), 43-67.
Antonis, B., Granstström, B., & Möbius, B. (2001). Developments and paradigms in
intonation research. Speech Communication, 33, 263-296.
Berk-Seligson, S. (2002). The bilingual courtroom (2nd ed.). London, Chicago: The
University of Chicago.
Bolinger, D. (1978). Intonation across languages. In: J.P. Greenberg, C.A. Ferguson &
E.A. Moravcsik (Eds.), Universals of human language, Vol. II Phonology (pp.
471-524). Stanford: Stanford University Press.
Bolinger, D. (1989). Intonation and its uses: Melody in grammar and discourse.
London, Melbourne, Auckland: Edward Arnold.
Bucholtz, M. (2000). The politics of transcription. Journal of Pragmatics, 32, 1439-
1465.
Castello, E. (2008). Text complexity and reading comprehension tests. Bern: Peter
Lang.
Chahal, D. (1999). A preliminary analysis of Lebanese Arabic intonation. In
Proceedings of the 1999 Conference of the Australian Linguistic Society.
Retrieved from http://www.als.asn.au/proceedings/als1999/chahal.pdf
Cook, N. (2002). Tone of voice and mind. Amsterdam and Philadelphia: John
Benjamins.
Couper-Khulen, E., & Selting, M. (1996). Introduction. In E. Couper-Khulen & M.
Selting (Eds.), Prosody in conversation: Interactional studies (pp. 1-10). New
York: Cambridge University Press.
Crystal, D. (1969). Prosodic systems and intonation in English. London and New
York: Cambridge University Press.
Crystal, D. (1997). A dictionary of linguistics and phonetics (4th ed.). Cambridge, MA:
Blackwell.
De Jong, K., & Zawaydeh, B.A. (1999). Stress, duration, and intonation in Arabic
word-level prosody. Journal of Phonetics, 27, 3-22.
Duranti, A. (1997). Linguistic Anthropology. Cambridge: Cambridge University Press.
Edwards, A.B. (1995). The practice of court interpreting. Amsterdam and
Philadelphia: John Benjamins.
Eggins, S. (2004). An introduction to systemic functional linguistics (2nd ed.). New
York and London: Continuum.
Esposito, N. (2001). From meaning to meaning: the influence of translation techniques
on non-English focus group research. Qualitative Health Research, 11(4), 568-
579.
Foulkes, P. & Docherty, G. (2006). The social life of phonetics and phonology.
Journal of Phonetics, 34, 409-438.
Fraser, H. (1996). Identifying evidentiary audio voices – what phonetic science can
and can’t do. Policing Issues and Practice Journal, 4(4), 39-43.
Fraser, H. (2003). Issues in transcription: factors affecting the reliability of transcripts
as evidence in legal cases. Forensic Linguistics,10(2), 203-226.
Fujisaki. H. (1983). Dynamic characteristics of voice fundamental frequency in
speech and singing. In P. MacNeilage (Ed.), The production of speech (pp. 39-
55). New York: Springer.

Translation  &  Interpreting  Vol  8  No  2  (2016) 61  

Gardner, R. (1994). Conversation analysis: Some thoughts on its applicability to
applied linguistics. In R. Gardner (Ed.), Spoken Interaction Studies in Australia.
Australian Review of Applied Linguistics, Series S, 11, (97-118).
Ghazali, S., Hamdi, R., & Barkat, M. (2002). Speech rhythm variation in Arabic
dialects. In Proceedings of Speech Prosody, Aix-en-Provence, 2002 (pp. 331-
334). Retrieved from http://sprosig.isle.illinois.edu/sp2002/pdf/ghazali-hamdi-
barkat.pdf
Gibbon, D. (1998). Intonation in German. In D. Hirst & A. Di Cristo (Eds.),
Intonation systems: A survey of twenty languages (pp. 78-95). Cambridge:
Cambridge University Press.
Grabe, E., Kochanski, G., & Coleman, J. (2005). The intonation of native accent
varieties in the British Isles - potential for miscommunication? In K. Dziubalska-
Kolaczyk & J. Przedlacka (Eds.), English pronunciation models: a changing
scene (pp. 311-337). Frankfurt am Main: Peter Lang.
Gumperz, J., & Gumperz. J.C. (1982). Introduction: language and the communication
of social identity. In J.J. Gumperz (Ed.), Language and social identity (pp. 1-21).
Cambridge: Cambridge University Press.
Gumperz, J.J. (1996). Foreword. In E. Couper-Khulen & M. Selting (Eds.), Prosody
in conversation: Interactional studies (pp. x-xii). New York: Cambridge
University Press.
Gumperz, J.J., & Roberts, C. (1991). Understanding the intercultural encounters. In J.
Blommaert & J. Verschueren (Eds.), The pragmatics of intercultural and
international communications (pp. 51-91). Amsterdam and Philadelphia: John
Benjamins.
Gupta, A.F. (1997). Colonisation, migration and functions of English. In E.W.
Schneider (Ed.), Englishes around the World 1: General studies, British Isles,
North America studies in honour of Manfred Görlach (pp. 47-58). Amsterdam
and Philadelphia: John Benjamins.
Gussenhoven, C. (2004). The phonology of tone and intonation. Cambridge:
Cambridge University Press.
Hale, S. (1996). Pragmatic considerations in court interpreting. Australian Review of
Applied Linguistics, 19(1), 61-72.
Hale, S. (1997). The treatment of register in court interpreting. The Translator:
Studies in Intercultural Communication, 3(1), 39-54.
Hale, S. (2002). How faithfully do court interpreters render the style of non-English
speaking witnesses’ testimonies? A data-based study of Spanish-English
bilingual proceedings. Discourse Studies, 4(1), 25-47.
Halliday, M.A.K., & Hasan, R. (1976). Cohesion in English. London: Longman.
Halliday, M.A.K. (1987). Spoken and written modes of meaning. In R. Horowitz & J.
Samuels (Eds.), Comprehending oral and written language (pp. 55-82). San
Diego and London: Academic Press.
Halliday, M.A.K. (1994). An introduction to functional grammar (2nd ed.). London:
Edward Arnold.
Holes, C. (1995). Modern Arabic: Structures, functions and varieties. London and
New York: Longman.
Horowitz, R., & Samuels, J. (1987). Comprehending oral and written language:
critical contrasts for literacy and schooling. In R. Horowitz & J. Samuels (Eds.),
Comprehending oral and written language (pp. 1-52). San Diego and London:
Academic Press.
House, J. (2001). Translation quality assessment: Linguistic description versus social
evaluation. Meta XLVI(2), 243-257.
Jun, S-A. (2005). Prosodic typology. In S-A. Jun (Ed.), Prosodic typology: The
phonology of intonation and phrasing (pp. 430-458). Oxford: Oxford University
Press.
Kulk, F., C. Odé, & Woidich, M. (2003). The intonation of colloquial damascene
Arabic: a pilot study. Proceedings 25, the Institute of Phonetic Sciences of the
University of Amsterdam, 15-20.
Ladd, D.R. (2001). Intonation. In M. Haspelmath, E. König, W. Oesterreicher, & W.
Raible (Eds.), Language typology and language universals: an international
handbook (pp. 1380-1390). Berlin and New York: de Gruyter.

Translation  &  Interpreting  Vol  8  No  2  (2016) 62  

Lieberman, P. (1967). Intonation, perception, and language. Cambridge, MA: MIT
Press.
Local, J. (1996). Conversational phonetics: some aspects of news receipts in everyday
talk. In E. Couper-Khulen & M. Selting (Eds.), Prosody in Conversation:
Interactional studies (pp. 177-230). New York: Cambridge University Press.
Merlini, R., & Favaron, R. (2005). The voice of interpreting in speech pathology.
Interpreting 7(2), 263-302.
Müller, E.F. (1996). Affiliating and disaffiliating with continuers: prosodic aspects of
recipiency. In E. Couper-Khulen & M. Selting (Eds.), Prosody in conversation:
Interactional studies (pp. 131-176). New York: Cambridge University Press.
Newmark, P. (1988). A Textbook of Translation. UK: Prentice Hall International.
Nida, E. (1964). Toward a Science of Translation. Leiden: E. J. Brill.
Odisho, E. (2005). Techniques of teaching comparative pronunciation in Arabic and
English. New Jersey: Gorgias Press.
O’Connor, J.D., & Arnold, G.F. (1973). Intonation of Colloquial English: A practical
handbook. London: Longman.
Pierrehumbert, J., & Hirschberg. J. (1990). The meaning of intonation contours in the
interpretation of discourse. In P.R. Cohen, J. Morgan, & M.E. Pollack (Eds.),
Intentions in communication (pp. 271-311) Cambridge, MA: MIT Press.
Prevignano, C.L., & di Luzio, A. (2003). A discussion with John J. Gumperz. In S. L.
Eerdmans, C. L. Prevignano, & P.J. Thibault (Eds.), Language and interaction:
Discussion with John J. Gumperz (pp. 7-29). Amsterdam and Philadelphia: John
Benjamins.
Rosenhouse, J. (1998). Women’s speech and language variation in Arabic dialects. Al-
‘Arabiyya 31, 123-152.
Sacks, H. (1995). Paraphrasing; alternative temporal references; approximate and
precise numbers; laughter; ‘uh huh’. In G. Jefferson (Ed.), Lectures on
Conversation: Volumes I and II (pp. 739-747). Oxford and Cambridge, MA:
Balckwell.
Shlesinger, M. (1994). Intonation in the production and perception of simultaneous
interpretation. In S. Lambert & B. Moser-Mercer (Eds.), Bridging the gap:
Empirical research in simultaneous interpretation (pp. 225-236). Amsterdam
and Philadelphia: John Benjamins.
Smirnova, N., Starshinov, A., Oparin, I., & Goloshchapova, T. (2007). Speaker
identification using selective comparison of pitch contour parameters. 16th
International Congress of Phonetic Sciences, 6-10 August, Saarbrucken,
Germany. Retrieved from http://icphs2007.de/conference/Papers/1138/1138.pdf
Tannen, D. (1982). Ethnic style in male-female conversation. In Gumperz, J. (Ed.),
Language and social interation (pp. 217-231). Cambridge: Cambridge
University Press.
Ten Have, P. (1999). Doing conversation analysis, a practical guide. London: Sage.
Teichman, D.E. (2000). Interpreting evidentiary tape recordings: the toughest job
you’ll ever love, or maybe not. Retrieved from http://www.linguisticworld.com/
documents/PDF%206%20ATA%20Chronicle%20article%20scan0001.pdf
Versteegh, K. (2001). The Arabic language. Edinburgh: Edinburgh University Press.
Watson, J.C.E. (2002). The phonology and morphology of Arabic. New York: Oxford
University Press.

Translation  &  Interpreting  Vol  8  No  2  (2016) 63